Access Fitbit Health and Fitness Data

Fitbit MCP is an unofficial Model Context Protocol implementation, built on Fitbit's public API, that lets an AI assistant access and analyze a user's Fitbit health and fitness data. It exposes sixteen tools: getUserProfile, getActivities, getSleepLogs, getHeartRate, getSteps, getBodyMeasurements, getFoodLogs, getWaterLogs, getLifetimeStats, getUserSettings, getFloorsClimbed, getDistance, getCalories, getActiveZoneMinutes, getDevices, and getBadges.

Use this connector when an assistant needs to answer questions about a user's Fitbit-tracked health and fitness data - recent activity, sleep quality, heart rate trends, calorie burn, or earned badges - conversationally instead of the user checking the Fitbit app themselves. It is explicitly unofficial and not affiliated with or endorsed by Fitbit Inc., built only against Fitbit's public API, so it is limited to whatever that API exposes and requires the user to have completed Fitbit's own OAuth2 flow to obtain an access token first.

Most tools accept an optional date parameter in YYYY-MM-DD format (defaulting to today) and a period parameter (1d, 7d, 30d, 1w, 1m) for time-ranged queries like steps, heart rate, or distance. Coverage spans daily activity metrics (steps, floors climbed, distance, calories, active zone minutes), body data (weight, body fat), sleep logs, food and water logs, lifetime statistics, connected device information, and earned badges/achievements.

{
  "command": "npx",
  "args": ["-y", "fitbit-mcp", "--stdio"],
  "env": { "FITBIT_ACCESS_TOKEN": "YOUR_FITBIT_ACCESS_TOKEN" }
}

or pass the token directly with --fitbit-token=YOUR_FITBIT_ACCESS_TOKEN instead of an environment variable. To obtain the token: register an application at the Fitbit Developer Portal with OAuth 2.0 Application Type "Personal" and callback URL http://localhost:3000, note the Client ID and Client Secret, then complete Fitbit's OAuth 2.0 authorization flow to receive an access token.

Users and developers who want an AI assistant to read and reason over their own Fitbit health data - reviewing sleep patterns, tracking activity or calorie trends over a period, checking heart rate history, or summarizing recent food and water logging - once they've completed Fitbit's OAuth flow to authorize access. It is a personal-use integration by design, since it targets Fitbit's "Personal" OAuth application type rather than a multi-user server-side deployment. The project is also open to contribution - it ships a standard npm-based development setup (clone, npm install, npm run build, npm run dev) and a GitHub Actions workflow that verifies every pull request's build and automates publishing new versions to npm on a tagged release.